A system and method for fraud detection and management are provided. The system includes a first communication device that receives a phone call from a second communication device, wherein a call flow of the phone call comprises one or more distinct phases. The system also includes a fraud detection and management system (FDMS) platform that determines whether the phone call exceeds a predetermined risk threshold at each distinct phase of the call flow.
Legal claims defining the scope of protection, as filed with the USPTO.
1. A fraud detection and management system (FDMS) that includes a computer or a server and that detects fraud in a call signal received from a caller device during one or more distinct phases of a call lifecycle, the fraud detection management system comprising: a controller that initiates and manages operations of a plurality of components in the fraud detection and management system (FDMS); a fraud detection module suite that includes a plurality of fraud detection modules, each of which employs a respective fraud detection technology based on the fraud or a phase of the call lifecycle; and a module manager that dynamically selects a fraud detection module from the plurality of fraud detection modules based on a phase of the call lifecycle and loads the selected fraud detection module into a platform process space to detect the fraud during said phase of the call lifecycle.
2. The system of claim 1 , wherein the fraud detection module suite comprises at least one of: a call detail record (CDR) analytics module that searches anomalous behaviors within a caller history; a voice print module that verifies a voice print of a caller voice signal against one or more voice prints stored in a database; a prosody module that analyzes prosodic features of an interaction between a caller on the caller device and a human agent on a caller agent device; and a predictive analytics module that predicts a fraud risk of the call signal.
3. The system of claim 2 , wherein the module manager selects the call detail record (CDR) analytics module to search anomalous behaviors within the caller history associated with metadata in the call signal.
4. The system of claim 2 , wherein the module manager selects the voice print module to verify a voice print of the caller voice signal against a plurality of voice prints stored in the database.
5. The system of claim 2 , wherein the module manager selects the prosody module to analyze prosodic features of the interaction between the caller on the caller device and the human agent on the caller agent device.
6. The system of claim 2 , wherein the module manager selects the predictive analytics module to predict the fraud risk of the call signal.
7. The system of claim 1 , wherein an initial risk estimate of fraud in the call signal is determined when the call signal is received, the initial risk estimate of fraud being based on at least one of a call number, a caller history associated with the call number, and a consistency of call metadata correlations of the caller history.
8. The system of claim 1 , wherein the system determines whether a recording of the call signal should be made based on a voice print of a caller when the caller interacts with an interactive voice response (IVR).
9. The system of claim 1 , wherein the plurality of components comprise a telnet management interface that facilitates loading, unloading, or upgrading one or more of the fraud detection modules.
10. The system of claim 1 , wherein the plurality of components comprise a notification manager that receives a fraud notification signal from the fraud detection module suite and dispatches a fraud alert signal to a client server.
11. The system of claim 1 , wherein the plurality of components comprise a publication/subscriber manager that receives the call signal and sends call data to the fraud detection module suite.
12. The system of claim 1 , wherein the plurality of components comprise a Representation State Transfer (REST) application program interface (API) server that includes a RESTful API that can be invoked by a client server to report detected fraud to the system.
13. The system of claim 1 , wherein the plurality of components comprise a case/machine learning manager that includes a plurality of case records that are referenced by the system to assess effectiveness of a current fraud model.
14. The system of claim 1 , wherein the controller comprises a boot strapper that instantiates each of the plurality of components.
15. The system of claim 1 , wherein the controller comprises a system monitor that monitors each of the plurality of components, and that receives a periodic status report signal from each of the plurality of components.
16. The system of claim 1 , wherein the controller comprises a configuration manager that loads and maintains an overall system configuration of the system.
17. A fraud detection and management system (FDMS) that includes a computer or a server and that detects fraud in a call signal received from a caller device during one or more distinct phases of a call lifecycle, the fraud detection management system comprising: a fraud detection module suite that includes a plurality of fraud detection modules, each of which employs a respective fraud detection technology based on the fraud or a phase of the call lifecycle; and a module manager that dynamically selects a fraud detection module from the plurality of fraud detection modules based on a phase of the call lifecycle and loads the selected fraud detection module into a platform process space to detect the fraud during said phase of the call lifecycle based on a client risk profile.
18. A method for fraud detection and management, the method comprising: receiving a call signal from a caller device; selectively identifying one or more fraud detection modules in a fraud detection module suite based on the call signal; dynamically loading the selectively identified one or more fraud detection modules in a fraud detection and management system platform process space; and processing the call signal by the loaded selectively identified one or more fraud detection modules in the fraud detection and management system platform process space to determine a fraud risk score for the call signal.
19. The method of claim 18 , further comprising: retrieving a client risk profile from a database; and processing the call signal by the selectively identified one or more fraud detection modules based on the client risk profile to determine the fraud risk score for the call signal.
20. The method of claim 18 , wherein the selectively identifying the one or more fraud detection modules is based on a stage of a lifecycle of the call signal, including before a call starts, after the call connects, during interactive voice response (IVR), or after connecting the call to a call center agent device.
Cooperative Patent Classification codes for this invention. Click any code to explore related patents in that topic.
July 25, 2017
April 14, 2020
Browse 5M+ US patents with plain-English claim translations and AI-generated analysis.